THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA PRESENTED BY PROF. CHARLES A. KOFOID AND MRS. PRUDENCE W. KOFOID 7 ON THE CLASSIFICATION AND GEOGRAPHICAL DISTRIBUTION OF THE MAMMALIA, BEING THE LECTURE ON SIR ROBERT READE'S FOUNDATION, DELIVERED BEFORE THE SSnftroftg of PRIMATES. ( Front teeth, piercers (6, 2, 10), laniaries i FER.E. S 1 TT 7 t J Front teeth, in both upper and lower jaw. BELLU^E. j t Front teeth, none in the upper jaw . . PECORA. ^ t Mutkate -Teeth variable CETE. On comparing the three preceding systems, it will be found that the most important errors of arrangement have been com- mitted, not by Aristotle, but by the modern naturalists. Both Bay and Linnaeus have mistaken the character of the horny parts enveloping the toes of the elephant, which do not defend the upper part merely, as is the case with claws, but embrace the under parts also, forming a complete case or hoof. With respect to Linnaeus, however, it must be observed, that although he has followed Bay in placing the elephant in the unguiculate group of quadrupeds, he has not overlooked the great natural divisions which the latter naturalist adopted from Aristotle; and his Ungulata is the more natural in the degree in which it approaches the corresponding group in the Aristotelian system. I now proceed to the arrangement of the Mammalia pro- posed by CuViER in .the last edition of his classical work entitled l Le Regne Animal distribue d'apres son organisation.'' Adopting the same threefold primary division of the class MAMMALIA as his predecessors, CUVIER subdivides it into 1 Aristotle knew that the Cetacea were mammiferous : 'TO,' (S 5&> /u.ti> Tobs) '5' &TOS, uffirep 5eA0is.' 2 From the Systema Nalurw, ed. xn. Holmiae, Tom. I. p. 24. more naturally defined orders, according to various characters afforded by the dental, osseous, generative and locomotive systems, which his great anatomical knowledge had made known to him. That heterogeneous order which Linnaeus prepossessed in favour of the easily recognisable outward character by which he distinguished the class had characterised by the 1 Mammce pectorales bince : dentes primores incisores : supe- riores i\ paralleli 1 ? was shewn, by the correlation of anatomi- cal distinctions with the threefold modification of the limbs of the Primates, to be divisible into as many distinct orders. The hands on the upper limbs alone, and the lower limbs destined to sustain the trunk erect, characterised the order Bimana, the equivalent of the Linnsean genus Homo. The genus Simia of Linnaeus, with hands on the four extremities, became the order Quadrumana of Cuvier. The genus Ves- pertilio with the 'manus palmatse volitantes' formed the group Cheiroptera, answerable to the Dermaptera of Aristotle. KAY had pointed out certain viviparous quadrupeds with a multifid foot as being " anomalous species," instancing as such " the tamandua, the armadillo, the sloth, the mole, the shrew, the hedgehog, and the bat." The first three species are associated with the scaly ant-eaters (Manis) of Asia and Africa, with the Australian spiny ant-eaters (Echidna], and with the more strange duck-moles (Ornithorhynchus) of the same part of the world, to form the order Edentata of Cuvier, which answers to that called Bruta by Linnaeus, if the ele- phant and walrus be removed from it. The rest of Rays anomalous species exemplify the families Cheiroptera and Insectivora of the Cuvierian system, in which they are asso- ciated with the true Carnivora in an order called ' Carnas- siers,' answering to the Ferce of Linnaeus. Cuvier had early noticed the relation of the Austra- lian pouched mammals, as a small collateral series, to the 1 Tom, cit. p. 24. 8 unguiculate mammals of the rest of the world; 'some,' he writes, l corresponding with the Carnivora, some with the Rodentia, and others again with the Edentata, by their teeth and the nature of their food.' They formed a family of the Carnassiers in the first edition of the ' Rtyne Animal 1 , but were raised to the rank of an order under the name Marsupialia in the second edition, where they terminate that series of the Unguiculata, which possess the three kinds of teeth incisors, canines and molars. The hoofed animals (UNGULATA, c animaux a sabots ') are binarily divided into those that do, and those that do not, chew the cud; the former constituting the order Pachyder- mata, the latter that of Ruminantia. The third primary group or subclass of Mammalia is indi- cated, but without receiving any name distinct from that of the single order Cetacea exemplifying it in the Cuvierian system an order which would be equivalent to the Mutica of the Linnaean system, save that the manatee which Linna3us placed in the same group as the elephant is associated with the whale in the Regne Animal. The Mammalian system of CuviER is exemplified in the subjoined Table: (See p. 9). Important as was the improvement it presented on previous classifications, the progress of anatomical and physiological knowledge, mainly stimulated by the writings and example of Cuvier himself, soon began to make felt the defects of his system. Shortly after its proposition, the zoological mind began to be disagreeably impressed by the results of the ap- plication of the characters employed by Cuvier in the forma- tion of the primary and secondary groups of the class ; the sloth, for example, being placed above the horse, the mole above the lynx, and the bat above the dog : even the Orni- thorhynchus paradoxus shewn by accurate anatomical scru- tiny to be the most reptilian of the mammalian class takes 1 8vo., 1816. 2 . .' r-i 5 10 precedence of the colossal and sagacious elephant in the Cu- vierian scheme 1 . The profound admiration and respect which I have always entertained for my chief instructor in Zootomy and Zoology, never blinded me to the necessity of much modification of his arrangement of the Mammalia. The question, more especi- ally, of the truly natural and equivalent primary groups of the class, has been present to my mind whenever I have been engaged in dissecting the rarer forms which have died at the Zoological Gardens in London, or on other occasions. But I propose first to submit to you, as briefly and clearly as I am able, the results of this store of anatomical knowledge as ap- plicable to the true organic characters of the class MAMMALIA. Mammals are distinguished outwardly by an entire or partial covering of hair 2 , and by having teats or mammae whence the name of the class. All mammals possess mammary glands and suckle their young : the embryo or foetus is developed in a womb. Their leading anatomical character is, the highly vascular and mi- Fig, i. nutely cellular structure of the lungs, (fig. 1, ,) which are freely suspended in a thoracic cavity separated by a musculo- tendinous partition or 'diaphragm' from the abdomen, (ib. d.) 1 The modifications consequently proposed by Geoffroy St Hilaire, Illiger, De Blainville, C. L. Bonaparte, J. E. Gray, Waterhouse, Milne Edwards, Lesson, Wagner, Nilsson, Oken, Macleay, Sir E. Home, Gervais, and others, have been cited and commented upon in my Papers communicated to the Lin- nsean Society (Proceedings, 1857) and the Geological Society (Proceedings, Nov. 1847, pp. 135140). 2 The foetal Cetacea shew tufts of hair on the muzzle. 11 Mammals, like Birds, have a heart composed of two ven- tricles and two auricles, and have warm blood : they breathe quickly ; but inspiration is performed chiefly by the agency of the diaphragm ; and the inspired air acts only on the capil- laries of the pulmonary circulation. The blood-discs are smaller than in Reptiles, and, save in the Camel-tribe, are circular in form. The right auriculo- ventricular valve is membranous, and the aorta bends over the left bronchial tube. The kidneys are relatively smaller and present a more compact figure than in the other vertebrate classes; their parenchyma is divided into a cortical and medullary portion, and the secreting tubuli terminate in a dilatation of the excre- tory duct, called the pelvis : they derive the material of their secretion from the arterial system. Their veins are simple, commencing by minute capillaries in the parenchyma and terminating generally by a single trunk on each side in the abdominal vena cava: they never anastomose with the mesen- teric veins. The liver is generally divided into a greater number of lobes than in Birds. The portal system is formed by veins derived exclusively from the spleen and chylopoietic viscera. The cystic duct, when it exists, always joins the hepatic, and does not enter the duodenum separately. The pancreatic duct is commonly single. The mouth is closed by soft flexible muscular lips: the upper jaw is composed of palatine, maxillary and premaxillary bones, and is fixed ; the lower jaw consists of two side-halves, or rami, which are simple or formed by one bony piece, and are articulated by a convex (fig. 3, b) or flat condyle to the base of the zygomatic process, and not to the tympanic ele- ment, of the temporal bone ; the base of the coronoid process (ib. c) generally extends along the space between the condy- loid and the alveolar processes. The jaws of Mammals with few exceptions are provided with teeth, which are arranged in a single row ; they are always lodged in sockets, and never 12 anchylosed with the substance of the jaw. The tongue is fleshy, well-developed, with the apex more or less free. The posterior nares are protected by a soft palate, and the larynx by an epiglottis : the rings of the trachea are generally carti- laginous and incomplete behind : there is no inferior larynx. The oesophagus is continued without partial dilatations to the stomach, which varies in its structure according to the nature of the food, or the quantity of nutriment to be extracted there- from. The trunk- vertebras of Mammalia have their bodies ossified from three centres, and present for a longer or shorter period of life a discoid epiphysis at each extremity. They are arti- culated by concentric ligaments with interposed glairy fluid forming what are called the intervertebral substances; the articulating surfaces are generally flattened, but sometimes, as in the necks of certain Ruminants, they are concave behind and convex in front. The cervical vertebras are seven in number, neither more nor less. The lumbar vertebras are more constant and usually more numerous than in other classes of vertebrate animals. The atlas is articulated by concave articular processes to two convex condyles, which are developed from the ex-occipital elements, or neurapo- physes, of the last cranial vertebra. The tympanic element of the temporal bone is restricted in function to the service of the organ of hearing, and never enters into the articula- tion of the lower jaw. The olfactory nerves escape from the cranial cavity through numerous foramina of a cribriform plate. The optic foramina are always distinct from one another. The scapula is generally an expanded plate of bone ; the coracoid, with two (monotrematous) exceptions, appears as a small process of the scapula. The sternum consists of a narrow and usually simple series of bones : the sternal por- tions of the ribs are generally cartilaginous and fixed to the vertebral portions without the interposition of a distinct arti- culation : there are no gristly or bony abdominal ribs or 13 abdominal sternum. The pubic and ischial arches are gene- rally complete, and united together by bony confluence on the sternal aspect, so that the interspace of the two pelvic arches is converted into two holes, called ' foramina obturatoria.' The sclerotic coat of the eye is a fibrous membrane, and never contains bony plates. In the quantity of aqueous humour and the convexity of the lens Mammals are generally intermediate between Birds and Fishes. The organ of hearing is characterized by the full development of the cochlea with a lamina spiralis : there are three distinct ossicles in the tympa- num ; the membrana tympani is generally concave externally ; the meatus auditorius externus often commences with a com- plicated external ear, having a distinct cartilaginous basis. The external apertures of the organ of smell are provided with moveable cartilages and muscles, and the extent of the internal organ is increased by accessory cavities or sinuses which communicate with the passages including the turbinated bones. There are few characters of the osseous system common, and at the same time peculiar, to the class Mammalia. The following may be cited : 1. Each half or ramus of the mandible consists of one bony piece developed from a single centre: the condyle is convex or flat, never concave. This has proved a valuable character in the determination of fossils. 2. The second or distal bone, called ' squamosal,' in the 'zygomatic' bar continued backward from the maxillary arch, is not only expanded, but is applied to the side-wall of the cranium, and developes the articular surface for the mandible, which surface is either concave or flat. 3. The presphenoid is developed from a centre distinct from that of the basisphenoid. In no other class of vertebrate animals are these osteo- logical characters present. The cancellous texture of mammalian Ibone is of a finer arid more delicate structure than in Keptiles, and forms a 14 closer network than in Birds. The microscopic radiating cells are relatively smaller and approach more nearly to the spheroid form. The Mammalia, like Eeptilia and Pisces, include a few genera and species that are devoid of teeth ; the true ant- eaters (Myrmecophaga), the scaly anteaters or pangolins .(Mam's), and the spiny monotrematous anteater (Echidna), are examples of strictly edentulous Mammals. The Orni- thorhynchus has horny teeth, and the whales (Balcena and Balcenoptera) have transitory embryonic calcified teeth, suc- ceeded by whalebone substitutes in the upper jaw. The female Narwhal seems to be edentulous, but has the germs of two tusks in the substance of the upper jaw-bones; one of these becomes developed into a large and conspicuous weapon in the male Narwhal, whence the name of its genus Monodon. The examples of excessive number of teeth are presented, in the order Bruta, by the priodont Armadillo, which has ninety-eight teeth : and in the Cetaceous order by the Cacha- lot, which has upwards of sixty teeth, though most of them are confined to the lower jaw; by the common Porpoise, which has between eighty and ninety teeth : by the Gangetic Dol- phin, which has one hundred and twenty teeth ; and by the true Dolphins (Delphinus] , which have from one hundred to one hundred and ninety teeth, yielding the maximum number in the class Mammalia. When the teeth are in excessive number, as in the Arma- dillos and Dolphins above cited, they are small, equal, or sub-equal, and usually of a simple conical form. In most other mammals particular teeth have special forms for special uses ; thus, the front teeth, (figs. 2 and 3, ,) from being commonly adapted to effect the first coarse division of the food, have been called cutters or incisors; and the back teeth, (ib. m,) which complete its comminution, grinders or molars; large conical pointed teeth situated behind the in- cisors, and adapted, by being nearer the insertion of the biting 15 muscles, to act with greater force, are called holders, tearers, laniaries, or more commonly canines, (ib. c,) from being well developed in the Dog and other Carnivora. It is peculiar to the class Mammalia to have teeth im- planted in sockets by two or more fangs ; but this can only happen to teeth of limited growth, and generally characterizes the molars and premolars : perpetually growing teeth require the base to be kept simple and widely excavated for the per- sistent pulp. In no mammiferous animal does anchylosis of the tooth with the jaw constitute a normal mode of attachment. Each tooth has its peculiar socket, to which it firmly adheres by the close co-adaptation of their opposed surfaces, and by the firm adhesion of the alveolar periosteum to the organized cement which invests the fang or fangs of the tooth. True teeth implanted in sockets are confined, in the Mam- malian class, to the maxillary, premaxillary, and mandibular or lower maxillary bones, and form a single row in each. They may project only from the premaxillary bones, as in the Narwhal ; or only from the lower maxillary bone, as in Ziphius; or be limited to the superior and inferior maxillaries and not present in the premaxillaries, as in the true Bumi- nantia and most Bruta (Sloths, Armadillos, Orycteropes). In most Mammals teeth are situated in all the bones above men- tioned. The teeth of the Mammalia usually consist of hard un- vascular dentine, defended at the crown by an investment of enamel, and everywhere surrounded by a coat of cement. The coronal cement is of extreme tenuity in Man, Quad- rumana and the terrestrial Carnivora; it is thicker in the Herbivora, especially in the complex grinders of the Ele- phant. Vertical folds of enamel and cement penetrate the crown of the tooth in the ruminating and many other Ungulata, and in most Eodents, characterizing by their various forms the genera of those orders. No Mammal has more than two sets of teeth. In some 16 species the tooth-matrix does not develope the germ of a second tooth, destined to succeed the one into which the matrix has been converted ; such a tooth, therefore, when completed and worn down, is not replaced. The Sperm Whales, Dol- phins, and Porpoises are limited to this simple provision of teeth. In the Armadillos and Sloths, the want of generative power, as it may be called, in the matrix is compensated by the persistence of the matrix, and by the uninterrupted growth' of the teeth. In most other Mammalia, the matrix of the first-developed tooth gives origin to the germ of a second tooth, which some- times displaces the first, sometimes takes its place by the side of the tooth from which it has originated. All those teeth which are displaced by their progeny are called 'temporary,' deciduous, or milk-teeth, (figs. 2 and 3, d, 1...4) ; the mode and direction in which they are displaced and succeeded, viz. from above downwards in the upper, from below upwards in the lower, jaw, in both jaws vertically are the same as in the Crocodile; but the process is never re- peated more than once in any mammalian animal. A con- siderable proportion of the dental series is thus changed ; the second or * permanent' teeth having a size and form as suitable to the jaws of the adult, as the ' temporary' teeth were adapted to those of the young animal. Those permanent teeth, which assume places not pre- viously occupied by deciduous ones, are always the most pos- terior in their position, and generally the most complex in their form. The term ' molar ' or ' true molar ' is restricted to these teeth (fig. 2 and 3, m). The teeth between them and the canines are called ' premolars,' (ib. p) ; they push out the milk-teeth, (ib. d,) and are usually of smaller size and simpler form than the true molars. Thus the class Mammalia, in regard to the times of form- ation and the succession of the teeth, may be divided into two groups, viz. Monophyodonts 1 or those that generate a s, once ; 0tfw, I generate ; oSovs, tooth. 17 single set of teeth, and the Diphyodonts 1 or those that generate two sets of teeth. But this dental character is not so asso- ciated with other organic characters as to indicate natural or equivalent sub-classes. In the Mammalian orders with two sets of teeth, these organs acquire individual characters, receive special denomi- nations, and can be determinated from species to species. This differentiation of the teeth is significative of the high grade of organization of the animals manifesting it. Originally, indeed, the names ' incisors,' e canines,' and ' molars,' were given to the teeth, in Man and certain Mam- mals, as in Keptiles and Fishes, in reference merely to the shape and offices indicated by those names ; but they are now used as arbitrary signs, in a more fixed and determinate sense. In some Garni vora, e. g., the front teeth have broad tubercu- late summits adapted for nipping and bruising, while the principal back-teeth are shaped for cutting and work upon each other like the blades of scissors. The front-teeth in the Elephant project from the upper jaw, in the form, size and direction of long pointed horns. Indeed, shape and size are the least constant of dental characters in the Mammalia ; and the homologous teeth are determined, like other parts, by their relative position, by their connexions, and by their development. Those teeth which are implanted in the premaxillary bones, and in the corresponding part of the lower jaw, are called 'incisors' (fig. 2, *'), whatever be their shape or size. The tooth in the maxillary bone, which is situated at or near to the suture with the premaxillary, is the ' canine,' as is also that tooth in the lower jaw (ib. c), which, in opposing it, passes in front of the upper one's crown when the mouth is closed. The other teeth of the first set are the ' deciduous molars ' (d. 1 3) ; the teeth which displace and succeed them vertically are the ' premolars ' (p. 1 3) ; the more posterior 1 Sis, twice ; . 4, together with m. 3, are more or less incomplete, and will be found concealed in their closed alveoli 1 . The last deciduous molar, J.4, has the same relative supe- riority of size to d. 3 and d. 2, which m. 3 bears to m. 2 and m. 1 ; and the crowns of p. 3 and p. 4 are of a more simple form than those of the milk-teeth, which they are destined to succeed. When the milk-teeth are shed, and the permanent ones are all in place, their kinds are indicated, in the genus Sus, by the following formula : . 33 1 1 44 33 *-^> c -r=T' p-4=i> -3^ =44 >' which signifies that there are on each side of both upper and lower jaws 3 incisors, 1 canine, 4 premolars, and 3 molars, making in all 44 teeth, each tooth being distinguished by its appropriate symbol, viz. p. 1 to p. 4, m. 1 to m. 3. This number of teeth is never surpassed in the placental diphyodont series. 1 I recommend this easily acquired ' subject' to the young zoologist for a demonstration of the most instructive peculiarities of the mammalian dentition. He will see that the premolars must displace deciduous molars in order to rise into place : the molars have no such relations. C 2 20 When the premolars and the molars are below this typical number, the absent teeth are missing from the back part of the molar series, and usually from the fore part of the pre- molar series. The most constant teeth are the fourth premolar and the first true molar. These being known by their order and mode of development, the homologies of the remaining molars and premolars are determined by counting the molars from before backwards, e. g. ' one,' ' two,' ' three,' and the pre- molars from behind forwards, l four,' ' three,' ' two,' ' one.' The incisors are counted from the median line, commonly the foremost part, of both upper and lower jaws, outwards and backwards. The first incisor of the right side is the homo- type, transversely, of the contiguous incisor of the left side in the same jaw, and vertically, of its opposing tooth in the opposite jaw ; and so with regard to the canines, premolars, and molars ; just as the right arm is the homotype of the left arm in its own segment, and also of the right leg of a suc- ceeding segment. It suffices, therefore, to reckon and name the teeth of one side of either jaw in a species with the typi- cal number and kinds of teeth, e. g. the first, second, and third incisors, the first, second, third, and fourth premolars, the first, second, and third molars ; and of one side of both jaws in any case. I have been induced to dwell thus long on the dental cha- racters of the class Mammalia, because they have not been rightly defined in any systematic or elementary work on zoplogy, although an accurate formula and notation of the teeth are of more use and value in characterizing genera in this than in any other class of animals. Mammals may be surpassed in the rapidity with which the blood circulates, in the extent and completeness of the respiratory processes,' in bodily temperature, in the concomi- tant vigour of the muscular actions ; all which superiorities, in Birds, for example, result in those marvellous powers of flight with which the feathered class is privileged. But in their psychical phenomena the Mammalia, as a class, excel all 21 other animals. Let me exemplify this by reference to the reproductive economy in the vertebrate series. The instinctive sense of dependence upon another, mani- fested by the impulse to seek out a mate, which impulse, even in fishes, is sometimes so irresistible that they throw themselves on shore in the pursuit, this first step in the supercession of the lower and more general law of individual- or self-preservation, although not first introduced at the ver- tebrate stage of the animal series, is never departed from after that stage has been gained. To this sexual relation is next added a self-sacrificing impulse of a higher kind, viz. the parental instinct. As we rise in the survey of vertebrate phenomena, we see the entire devotion of self to offspring in the patient incubation of the bird, in the unwearied exertions of the Swift or the Hawk to obtain food for their callow brood when hatched ; in the bold demonstration which the Hen, at other times so timid, will make to repel threatened attacks against her cowering young. Still closer becomes the link between the parent and off- spring in the Mammalian class, by the substitution, for the exclusion of a passive irresponsive ovum, of the birth of a living young, making instinctive irresistible appeal, as soon as born, to maternal sympathy ; deriving nutriment immedi- ately from the mother's body, and both giving and receiving pleasure by that act. These beautiful foreshadowings of higher attributes are, however, transitory in the brute creation, and the relations cease, as soon as the young quadruped can provide for itself. Preservation of offspring has been superinduced on self-pre- servation, but there is as yet no self-improvement : this is the peculiar attribute of mankind. The human species is charac- terised by the prolonged dependence of a slowly maturing off- spring on parental cares and affections, in which are laid the foundations of the social system, and time given for instilling those principles on which Man's best wisdom and truest hap- piness are based, and by which he is prepared for another and 22 a higher sphere of existence. In this destination alone may we discern an adequate end and purpose in the great organic scheme developed upon our planet. The progressive gradations in this scheme will be further exemplified as I proceed to explain the principles and cha- racters by which I have been guided in the formation of the primary groups or divisions of the class Mammalia. Prior to the year 1836 it was held by comparative ana- tomists that the brain in Mammalia differed from that in all other vertebrate animals by the presence of the large mass of transverse white fibres, called 'corpus callosum' by the anthropotomist; which fibres, overarching the ventricles and diverging as they penetrate the substance of either hemisphere of the cerebrum, bring every convolution of the one into com- munication with those of the other hemisphere, whence the other name of this part the * great commissure.' In that year I discovered that the brain of the kangaroo, the wombat, and some other marsupial quadrupeds, wanted the ' great commis- sure;' and that the cerebral hemispheres were connected together, as in birds, only by the 'fornix' and ' anterior commissure 1 .' Soon afterward, I had the opportunity of determining that the same deficiency of structure prevailed in the Ornithorhynchus and Echidna*. As many other modifications of structure, more or less akin to those characterizing birds and reptiles, were found to be associated with the above oviparous type of brain, together with some remarkable peculiarities in the economy of repro- duction 3 , I pointed out that the Mammalia might be divided into 'placental' and ' implacentalV Impressed, however, with the fact that such binary divi- sion, like that which might be based upon the leading differ- ences of dentition, was too unequal to be natural, the larger 1 See Philosophical Trans, for 1837, p. 87. 2 Art. MONOTREMATA, Cyclopcedia of Anatomy and Physiology, Vo\. 111^.383. 3 Art. MAESUPIALIA, torn. cit. p. 257. 4 Art. MAMMALIA, torn. cit. p. 244. 23 group never presenting the same degree of correspondence of organic structure as the smaller moiety, I continued to pursue investigations, with the view of gaming an insight into the more natural and equivalent primary groups of the Mammalia; having my attention more especially directed to the cerebral organ in this quest. In 1842, I was able to demonstrate, in the ' Hunterian Course of Lectures' delivered at the Royal College of Sur- geons, the leading modifications of the mammalian brain, and their peculiar value in classification by reason of their asso- ciation with concurrent modifications of other systems of organs. Nevertheless there were genera of Mammals, e. g. the sloths, anteaters, armadillos, roussettes, giraffes, rhinoceroses, &c. to which the cerebral test had to be applied. Fortunately the rare species of these genera successively arrived at the Zoological Gardens in London, and afforded me the means of applying that test ; so that, at length, having dissected the brain in one species at least, of almost every genus or natural family of the Mammalian class, I felt myself in a position to submit to the judgment of my fellow-labourers in zoology, at the Linnaean Society, in 1857, the generalised results of such dissections, comprising a fourfold primary divi- sion of the MAMMALIA, based upon the four leading modi- fications of cerebral structure in that class. In some mammals the cerebral hemispheres are but feebly and partially connected together by the 'fornix' and ' ante- rior commissure:' in the rest of the class the part called ' corpus callosum' is added, which completes the connecting or ' commissural ' apparatus. With the absence of this great superadded commissure 1 is associated a remarkable modification of the mode of develop- ment of the offspring, which involves many other modifica- tions; amongst which are the presence of the bones called { marsupial,' and the non-development of the deciduous body 1 On t,7ie Structure of the Brain in Marsupial Animals, PkiJos. Tram. 1837. 24 concerned in the nourishment of the progeny before birth, called t placenta;' the young in all this ' implacental ' divi- sion being brought forth prematurely, as compared with the rest of the class. This first and lowest primary group, or subclass, of Mam- malia is termed, from its cerebral character, LYENCEPHALA I , signifying the comparatively loose or disconnected state of the cerebral hemispheres. The size of these hemispheres (fig. 4, A) is so small that they leave exposed the olfactory gan- glions (a), the cerebellum (c), and more or less of the optic lobes (B) ; their surface is generally smooth ; the anfractuosi- ties, when present, are few and simple. Fig- 5- Fig. 4 . Brain of Opossum. Brain of Beaver. The next well marked stage in the development of the brain is where the corpus callosum (indicated in fig. 5, by the dotted lines d, d) is present, but connects cerebral hemispheres as little advanced in bulk or outward character as in the pre- ceding subclass ; the cerebrum (A) leaving both the olfactory lobes (a) and cerebellum (c) exposed, and being commonly tfw, to loose ; , brain. 25 smooth, or with few and simple convolutions in a very small proportion, composed of the largest members, of the group. The mammals so characterized constitute the subclass LISSEN- CEPHALA* (fig. 5). The third leading modification of the Mammalian cere- brum is such an increase in its relative size, that it extends over more or less of the cerebellum ; and generally more or less over the olfactory lobes. Save in very few exceptional cases of the smaller and inferior forms of Quadrumana (fig. 6), the superficies is folded into more or less numerous gyri or convolutions (fig. 7), whence the name GYRENCEPHALA, which I propose for the third subclass of Mammalia 2 . Fig. 7. Fig. 6. Brain of Marmoset Monkey. (Nat. size). Brain of Chimpanzee. (Half nat. size). In Man the brain presents an ascensive step in develop- ment, higher and more strongly marked than that by which the preceding subclass was distinguished from the one below it. Not only do the cerebral hemispheres overlap the olfac- tory lobes and cerebellum, but they extend in advance of the 1 \i<7a\os, brain. 26 one and further back than the other (figs. 8 & 9). Their pos- terior development is so marked that anthropotomists have assigned to that part the character and name of a ' third lobe :' it is peculiar and common to the genus Homo : equally pecu- liar is the ' posterior horn of the lateral ventricle' and the 'hippocampus minor,' which characterize the hind lobe of each hemisphere. Fig . 8 The superficial grey matter of the cere- brum, through the number and depth of the convolutions, attains its maximum of extent in Man. Peculiar mental powers are associ- ated with this high- est form of brain, and their consequences wonderfully illus- trate the value of the cerebral character ; according to my es- timate of which, I am led to regard the genus Homo as not merely a representa- tive of a distinct or- der, but of a distinct subclass,oftheMam- malia, for which I propose the name of ARCHENCEPHALA (tig. 9) . Ib. Side view, one-third nat. size. Brain of Negro, upper view. Fig. 9. , to overrule ; fy*l0aXos, brain. 27 With this preliminary definition of the organic characters, which appear to guide to a conception of the most natural primary groups of the class MAMMALIA, I next proceed to define the groups of secondary importance, or the subdivisions of the foregoing subclasses. The Lyencephalous Mammalia are unguiculate : some have the 'optic lobes' simple, others partly subdivided, or complicated by accessory ganglions, the lobes being then called 'bigeminal bodies.' The LYENCEPHALA with simple optic lobes are ' edentulous ' or without calcified teeth, and are devoid of external ears, scrotum, nipples, and marsupial pouch : they are true ' testi- conda :' they have a coracoid bone extending from the scapula to the sternum, and also an epicoracoid and epi- sternum as in Lizards : they are unguiculate and pentadactyle, with a supplementary tarsal bone supporting a perforated spur in the male. The order so characterized is called 1 MONOTREMATA,' in reference to the single excretory and generative outlet, which, however, is by no means peculiar to them among Mammalia. It includes two genera Echidna and Ornithorhynchus. Of the first, the species are terrestrial, insectivorous, chiefly myrmecophagous, having the beak-like slender jaws, and long cylindrical tongue of the true anteaters ; but they are covered, like the hedgehog, with spines. Of the second genus, the species are aquatic, with a flattened beak, like that of a duck, which is used in the anserine manner to extract insects and worms from the mud : but they are clothed with a close fine fur like that of a mole, whence the name * duck-mole' by which these anomalous quadrupeds are com- monly known to the colonists. Both genera of Monotremes are strictly limited to Australia and Tasmania. The LYENCEPHELA with divided optic lobes, forming the 'corpora bigemina' and 'quadrigemina' of anthropotomists, have teeth, and with rare exceptions, the three kinds, viz. incisors, canines, and molars. They are called MARSUPIA- LIA, because they are distinguished by a peculiar pouch or 28 duplicature of the abdominal integument, which in the males is everted, forming a pendulous bag, arid in the females is inverted, forming a hidden pouch containing the nipples and usually sheltering the young for a certain period after their birth : they have the marsupial bones in common with the Monotremes ; a much varied dentition, especially as regards the number of incisors, but usually including 4 true molars ; and never more than 3 premolars 1 (fig. 2) : the angle of the lower jaw (ib. a) is more or less inverted 2 . With the exception of one genus, Didelphys, which is Ame- rican, and another genus Cuscus, which is Malayan, all the known existing Marsupials belong to Australia, Tasmania, and New Guinea. The grazing and browsing Kangaroos are rarely seen abroad in full daylight, save in dark rainy weather. Most of the Marsupialia are nocturnal. Zoological wanderers in Australia, viewing its plains and scanning its scrubs by broad daylight, are struck by the seeming absence of mam- malian life ; but during the brief twilight and dawn, or by the light of the moon, numerous forms are seen to emerge from their hiding-places and illustrate the variety of marsupial life with which many parts of the continent abound. We may associate with their low position in the mammalian scale the prevalent habit amongst the Marsupialia of limiting the exer- cise of the faculties of active life to the period when they are shielded by the obscurity of night. The premature birth of the offspring, and its transference to the tegumentary pouch, in whicli it remains suspended to the nipple for a period answering to that of uterine life in higher mammals, relate to the peculiarities of the climate of Australia. The adventurous and much-enduring explorers of that continent bear uniform testimony to the want of water as the 1 Outlines of a Classification of the Marsupialia, Trans. Zool.Soc. Vol. n. 1839. 2 For other Osteological and Dental characteristics of the Marsupialia, see the paper above cited, and that On the Osteology of the Marsupialia, Trans. Zool. Soc. Vol. n. p. 379 (1838). 29 chief cause of their sufferings and danger. During the dry season the rivers are converted into pools, ' few and far be- tween ;' and the drought is sometimes continued so long as to dry up these. An ordinary non-marsupial quadruped, such as the wild cat or fox, having deposited her young in the nest or burrow, would in such a climate, at the droughtiest period of her existence, be compelled to travel a hundred, perhaps two hundred miles, in order to quench her thirst. Before she could return her blind and helpless litter would have perished. By the marsupial modification the mother is enabled to carry her offspring with her in the long migrations necessitated by the scarcity of water. With the climatal peculiarities of Australia, therefore, we may connect the peculiar modifications of those members of the mammalian class which are most widely distributed over that continent. But the principle of final causes receives more especial illustrations from the contingent particulars of the marsupial organization. The new-born Kangaroo is an inch in length, naked, blind, with very rudimental limbs and tail : in one which I examined the morning after the birth, I could discern no act of sucking : it hung, like a germ, from the end of the long nipple, and seemed unable to draw sustenance therefrom by its own efforts. The mother, accordingly, is provided with a peculiar adaptation of a muscle (cremaster) to the mammary gland, by which she can inject the milk from the nipple into the mouth of the pendulous embryo. Were the larynx of the little creature like that of the parent, the milk might probably would enter the windpipe and cause suffocation : but the foetal larynx is cone-shaped, with the opening at the apex, which projects, as in the whale-tribe, into the back aperture of the nostrils, where it is closely em- braced by the muscles of the ' soft palate.' The air-passage is thus completely separated from the fauces, and the injected milk passes in a divided stream on either side the base of the larynx into the oesophagus. These correlated modifica- tions of maternal and foetal structures, designed with especial reference to the peculiar conditions of both mother and off- spring, afford, as it seems to me, irrefragable evidence of Creative foresight. The LISSENCEPHALA, or smooth-brained placental Mam- malia, form a group which I consider as equivalent to the LYENCEPHALA or Implacentals ; and which includes the following orders, Rodentfa, Insectivora, Cheiroptera and Bruta. The RODENTIA are characterized by two large and long- curved incisors in each jaw, separated by a wide interval from the molars ; the teeth being so constructed, and the jaw so articulated, as to effect the reduction of the food to small particles by acts of rapid and. continued gnawing, whence the name of the order. The orbits are not separated from the temporal fossa?. The male glands pass periodically from the abdomen into a temporary scrotum, and are asso- ciated with prostatic and vesicular glands. The placenta is commonly discoid, but is sometimes a circular mass (Cavy), or flattened and divided into three or more lobes (Lepus). The Beaver and Capybara are the giants of the order, which chiefly consists of small, numerous, prolific and diversified unguiculate genera, subsisting wholly or in part on vegetable food. Some Rodents, e.g. the Lemmings, perform remark- able migrations, the impulse to which, unchecked by dangers or any surmountable obstacles, seems to be mechanical. Many Rodents build very artificial nests, and a few manifest their constructive instinct in association. In all these inferior psychical manifestations we are reminded of Birds. Many Rodents hibernate like Reptiles. They are distributed over all continents. About two-thirds of the known species of Mammalia belong to the Rodent order. The transition from the Marsupials to the Rodents is made by the Wombats ; and a transition from the Marsupials is made, by an equally easy step, through the smaller Opos- sums to the INSECTIVORA. This term is given to the order of small smooth-brained Mammals, the molar teeth of which are 31 bristled with cusps, and are associated with canines and incisors : they are unguiculate, plantigrade, and pentadactyle, and they have complete clavicles. Like Rodents, they are temporary testiconda, and have large prostatic and vesicular glands : like most other Lissencephala, the Insectivora have a discoid or cup-shaped placenta. They do not exist in South America and Australia ; their office in these continents is fulfilled by Marsupialia; but true Insectivora abound in all the other continents and their contiguous islands. The order CHEIROPTERA, with the exception of the modi- fication of their digits for supporting the large webs that serve as wings, repeat the chief characters of the Insectivora: a few, however, of the larger species are frugivorous and have corresponding modifications of the teeth and stomach. The mammas are pectoral in position. The most remarkable examples of periodically torpid Mammals are to be found in the terrestrial and volant Insecti- vora. The frugivorous Bats differ much in dentition from the true Cheiroptera, and would seem to conduct through the Colugos or Flying Lemurs, directly to the Quadrumanous order. The Cheiroptera are cosmopolitan. The order BRUTA, called Edentata by Cuvier, includes two genera (Myrmecophaga and Manis) which are devoid of teeth ; the rest possess those organs, which, however, have no true enamel, are never displaced by a second series, and are very rarely implanted in the premaxillary bones. All the species have very long and strong claws. The ischium as well as the ilium unites with the sacrum ; the orbit is not divided from the temporal fossa. The Three-toed Sloths (Brady- pus) manifest their affinity to the oviparous Yertebrata by the supernumerary cervical vertebras supporting false ribs and by the convolution of the wind-pipe in the thorax ; and I may add that the unusual number three and twenty pairs of ribs, forming a very long dorsal, with a short lumbar, region of the spine, in the Two-toed Sloth (Cholcepus), recalls a lacertine structure. The same tendency to an inferior type 32 is shown by the abdominal testes, the single cloacal outlet, the low cerebral development, the absence of medullary canals in the long bones in the Sloths, and by the great tenacity of life and long-enduring irritability of the muscular fibre, in both the Sloths and Anteaters 1 . The order Bruta is but scantily represented at the present period. One genus, Manis or Pangolin, is common to Asia and Africa; the Orycteropus is peculiar to South Africa; the rest of the order, consisting of the genera Myrmecophaga, or true Anteaters, Dasypus or Armadillos, and Bradypus or Sloths, are confined to South America. Having defined the orders or subdivisions of the two fore- going subclasses, I may remark that the LYENCEPHALA cannot be regarded as equivalent merely to one of the orders, say Rodentia, of the LISSENCEPHALA, without undervaluing the anatomical characters which are so remarkable and distinct in the marsupial and monotrematous animals. The anato- mical peculiarities of the edentulous LYENCEPHALA 2 appear to me to be, at least, of ordinal importance. In these deduc- tions I hold the mean between those who, with Geoffrey St Hilaire, would make a distinct class of the Monotremata, and those who, with Cuvier, would make the Monotremes a mere family of the Edentata. In like manner, whilst I regard the LYENCEPHALA as forming a group of higher rank than an order, I do not consider it as forming an equivalent primary group to that formed by all the placental Mammalia. The true value of the LYENCEPHALA is that of one of four 1 This latter vital character attracted the notice of the earliest observers of these animals. Thus Marcgrave and Piso narrate of the Sloth : ' Cor mo turn suum validissime retinebat, postquam exemptum erat e corpore per semiho- rium : exempto corde cseteris visceribus multo post se movebat et pedes lente contrahebat sicut dormituriens solet.' Buffon, who quotes the above from the Historia Naturalis Brasilia, p. 322, well remarks, ' Par ces rapports, ce quad- rupede se rapproche non seulement de la tortue, dont il a la lenteur, mais en- core des autres reptiles et de tous ceux qui n'ont pas un centre du sentiment unique et bien distinct.' Hist. Naturelle, 4to, Tom. xm. p. 45. 2 See my article Monotremata, in the Cyclopaedia of Anatomy, part xxvi. 1841. 33 primary divisions or subclasses of the Mammalia; its true equivalency is with the LISSENCEPHALA, and all its analogical relations are to be found more directly in that smooth-brained subclass than in the Placentalia at large. The following Table exemplifies the correspondence of the groups in the Lyencephalous and Lissencephalous series : LYENCEPHALA. LISSENCEPHALA. Rhizophaga ' Burrowing Rodentia. Poephaga l Dipodidce and Leporidce. Petaurus Pteromys. Phalangistidce Sciuridce and prehensile-tailed arboreal Rodents. Phascolarctos Bradypus. Perameles and Myrmecobius. Erinaceidce. Chceropus Macroscelis. Diddphys and PJiascogale . Soricidce. Dasyuridce Centetes, Gymnura. Echidna Manis. Besides the more general characters by which the LISSEN- CEPHALA, in common with the LYENCEPHALA, resemble Birds and Reptiles, there are many other remarkable indications of their affinity to the Oviparous Vertebrata in particular orders or genera of the subclass. Such, e. g., are the cloaca, con- voluted trachea, supernumerary cervical vertebras and their floating ribs, in the three-toed Sloth ; the numerous trunk-ribs in the two-toed Sloth ; the irritability of the muscular fibre, and persistence of contractile power in the Sloths and some other Bruta; the long, slender, beak-like edentulous jaws and gizzard of the Anteaters ; the imbricated scales of the equally edentulous Pangolins, which have both gizzard and gastric glands like the proventricular ones in birds ; the dermal bony armour of the Armadillos like that of loricated Saurians ; the quills of the Porcupine and Hedgehog ; the brilliant iridescent colours of the fur of the Cape-mole ( Clirysochlora aurea) ; the proventriculus of the Dormouse and Beaver; the pre- 1 On the Classification of the Marsupialia, Trans, of the Zool. Soc. Vol. n. P- 3i5> l8 39- D 34 valence of disproportionate development of the hind limbs in the Rodentia, coupled, in the Jerboa, with confluence of the three chief metatarsals into oiie bone, as in birds ; the keeled sternum and wings of the Bats ; the aptitude of the Cheirop- tera, Insectivora, and certain Rodentia to fall, like Keptiles, into a state of true torpidity, associated with a corresponding faculty of the heart to circulate carbonized or black blood : these, and the like indications of coaffinity with the LYEN- CEPHALA to the Oviparous air-breathing Yertebrata, have mainly prevailed with me against an acquiescence in the elevation of different groups of the LISSENCEPHALA to a higher place in the Mammalian series, and in their respective associa- tion, through some single character, with better-brained orders, according to Mammalogical systems which, at different times, have been proposed by zoologists of deserved reputation. Such, e. g., as the association of the long-clawed Bruta with the Ungulata 1 , and of the shorter-clawed Shrews, Moles and' Hedgehogs, as well as the Bats, with the Carnivora 2 ; of the Sloths with the Quadrumana 3 ; of the Bats with the same high order 4 ; and of the Insectivora and Rodentia in immediate sequence after the Linnean 'Primates,' as in the latest pub- lished < System of Mammalogy,' from a distinguished French author 5 . So far as their ordinal affinities are known, the most ancient Mammals, the fossil remains of which have been found in secondary strata, are either ly- or liss-encephalous, and belong either to the Marsupialia or the Insectivora. (Appendix A). In the GYRENCEPHALA WQ look in vain for those marks of affinity to the oviparous vertebrate animals which have been in- stanced in the preceding subclasses ; although, it is true, that when we proceed to consider the subdivisions of the GYREN- 1 Macleay, Linn. Trans. Vol. xvi. (1833) ; Gray, Dr. J. E., Mammalia in the British Museum, i2mo, 1843, p. xii. 2 Cuvier, R^gne Animal, 1829, p. 'no. 3 De Blainville, Osteographie, 4to, Fasc. i. p. 47 (1839). 4 Linnaeus, Systema Natures, Ed. 12, Tom. I. p. 26. 5 Prof. Gervais, Zoologie et PaUontologie Franpais, 4to, 1852, p. 194. 35 y we seem at first to descend in the scale by finding in that wave-brained subclass a group of animals, having the form of Fishes : but a high grade of mammalian organization is masked beneath this form. The GYRENCEPHALA are primarily subdivided, according to modifications of the locomotive organs, into three series, for which the Linnean terms may well be retained ; viz. Mutilata, Ungulata and Unguiculata, the maimed, the hoofed, and the clawed series. These limb-characters can only be rightly applied to the gyrencephalous subclass ; they do not indicate natural groups, save in that section of the Mammalia. To associate the LYENCEPHALA and LISSENCEPHALA with the unguiculate GYREN- CEPHALA into one great primary group, as in the Mammalian systems of Ray, Linnasus and Cuvier, is a misapplication of a solitary character akin to that which would have founded a primary division on the discoid placenta or the diphyodont dentition. No one has proposed to associate" the unguiculate Bird or Lizard with the unguiculate Ape; and it is but a little less violation of natural affinities to associate the Mono- tremes with the Quadrumanes in the same primary (unguicu- late) division of the Mammalian class. The three primary divisions of the GYRENCEPHALA are of higher value than the ordinal divisions of the LISSENCEPHALA ; just as those orders are of higher value than the representative families of the LYENCEPHALA. The Mutilata, or the maimed Mammals with folded brains, are so called because their hind limbs seem, as it were, to have been amputated ; they possess only the pectoral pair of limbs, and these in the form of fins : the hind end of the trunk expands into a broad, horizontally flattened, caudal fin. They have large brains with many and deep convolutions, are naked, and have neither neck, scrotum, nor external ears. The first order, called CETACEA, in this division are either edentulous or monophyodont, and the latter have teeth of one kind and usually of simple form. They are c testiconda,' and D2 36 have no 'vesiculse seminales.' The mammae are pudendal; the placenta is diffused; the external nostrils single or double are on the top of the head, and called spiracles or 'blow-holes.' They are marine, and, for the most part, range the unfathomable ocean; though with certain geogra- phical limits as respects species. The ' right whale ' of the northern hemisphere (Balcena mysticetus) is represented by a distinct species (Balo&na australis) in the southern hemi- sphere : the high temperature of the waters at the equatorial zone bars the migration of either from one pole to the other. True Cetacea feed on fishes or marine animals. The second order, called SIEENIA, have teeth of different kinds, incisors which are preceded by milk-teeth, and molars with flattened or ridged crowns, adapted for vegetable food. The nostrils are two, situated at the upper part of the snout ; the lips are beset with stiff bristles ; the mamma3 are pectoral ; they are ' testiconda,' but have 'vesiculae seminales.' The Sirenia exist near coasts or ascend large rivers ; browsing on fuci, water plants, or the 1 grass of the shore. There is much in the organization of this order that indicates its nearer affinity to members of the succeeding division, than to the cetaceous order. The Dugongs (Halicore) inhabit the Eed sea, the Ma- layan Archipelago, and the soundings of the Australian coasts : the Manatees (Manatus) frequent the shores of tropi- cal America and Africa. In the Ungulata the four limbs are present, but that por- tion of the toe which touches the ground is incased in a hoof, which blunts its sensibility and deprives the foot of prehen- sile power. With the limbs restricted to support and loco- motion, the Ungulates have no clavicles; the two bones of the fore leg are fixed together in the position anatomists call 'prone;' as a general rule hoofed quadrupeds feed on vegetables. A particular order, or suborder, of this group is indicated by fossil remains of certain South American genera, e. g. Toxodon and Nesodon, with long, curved, rootless teeth, 37 having a partial investment of enamel, and with certain pecu- liarities of cranial structure : the name TOXODONTIA is pro- posed for this order, all the representatives of which are extinct l . A second remarkable order, most of the members of which have also passed away, is characterized by two incisors in the form of long tusks ; in one genus (Dinotherium) projecting from the under jaw, in another genus (Elephas) from the upper jaw, and in some of the species of a third genus (Masto- don) from both jaws. There are no canines : the molars are few, large and transversely ridged ; the ridges sometimes few and mammillate, often numerous and with every intermediate gradation. The nose is prolonged into a cylindrical trunk, flexible in all directions, highly sensitive, and terminated by a prehensile appendage like a finger : from this peculiar organ is derived the name PEOBOSCIDIA given to the order. The feet are pentadactyle, but the toes are indicated only by divisions of the hoof; the placenta is annular ; the mammse are pectoral. Elephants are dependent chiefly upon trees for food. ' One species now finds the conditions of its existence in the rich forests of tropical Asia ; a second species in those of tropical Africa. Why, we may ask, should not a third be living at the expense of the still more luxuriant vegetation watered by the- Oronoko, the Essequibo, the Amazon, and the La Plata, in tropical America ? Geology tells us that at least two kinds of Elephant (Mastodon Andium and Mast. Humboldtii) for-, merly did derive their subsistence, along with the great Mega- therioid beasts, from that abundant source : two other kinds of Elephant (Mastodon ohioticus and Elephas texianus) existed in the warm and temperate latitudes of North America. Twice as many species of Mastodon and Elephant, distinct from all the others, roamed in pliocene times in the same latitudes of Europe. At a later or pleistocene period, a huge elephant, clothed with wool and hair, obtained its food from hardy trees, such as now grow in the 65th degree of north latitude ; and 1 Philosophical Transactions, 1853, p. 291. 38 abundant remains of this Elephas primigenius (as it has been prematurely called, since it was the last of our British ele- phants) have been found in temperate and high northern lati- tudes in Europe, Asia and America. This, like other Arctic animals, was peculiar in its family for its range in longitude. The Musk Buffalo was its contemporary in England and Eu- rope, and still lingers in the northernmost parts of America. I have received evidences of Elephantine species from China and Australia, proving the proboscidian pachyderms to have once been the most cosmopolitan of hoofed herbivorous quadrupeds. Both the proboscidian and toxodontal orders of UNGULATA may be called aberrant : the dentition of the latter, and several particulars of the organization of the Elephant, indicate an affinity to the Rodentia ;. the cranium of the Toxodon, like that of the Dinothere, resembles that of the Sirenia in its re- markable modifications. The typical Ungulate quadrupeds are divided, according to the odd or even number of the toes, into PERTSSODACTYLA and ARTIODACTYLA * : the single hoof of the horse, the triple hoof of the tapir, exemplify the first : the double hoof of the camel, the quadruple hoof of the hippopotamus, exemplify the second. In the perissodactyle or odd-toed UNGULATA, the dorso- lumbar vertebrae differ in number in different species, but are never fewer than twenty-two; the femur has a third trochanter, and the medullary artery does not penetrate the fore part of its shaft. The fore part of the astragalus is divided into two very unequal facets. The os magnum and the digitus medius which it supports are large, in some disproportionately so, and the digit is symmetrical : the same applies to the ectocunei- form and the digit which it supports in the hind foot. If the species be horned, as the Rhinoceros, the horn is single ; or, if there be two, they are placed on the median line of the head, one behind the other, each being thus an odd horn. 1 From 7repi